Data loading part

import pandas as pd

city_data_df = pd.read_csv ("city_data.csv")
city_list_df = pd.read_csv ("city_list.csv")
global_data_df = pd.read_csv ("global_data.csv")

print(city_data_df.count())

year        70792
city        70792
country     70792
avg_temp    68245
dtype: int64

city_data_df.describe()

	year	avg_temp
count	70792.000000	68245.000000
mean	1905.643576	16.138845
std	67.187043	7.603714
min	1743.000000	-14.350000
25%	1858.000000	9.710000
50%	1910.000000	16.080000
75%	1962.000000	23.820000
max	2013.000000	30.730000

print(city_data_df.head())

   year     city        country  avg_temp
0  1849  Abidjan  Côte D'Ivoire     25.58
1  1850  Abidjan  Côte D'Ivoire     25.52
2  1851  Abidjan  Côte D'Ivoire     25.67
3  1852  Abidjan  Côte D'Ivoire       NaN
4  1853  Abidjan  Côte D'Ivoire       NaN

print(city_list_df.head())

        city               country
0    Abidjan         Côte D'Ivoire
1  Abu Dhabi  United Arab Emirates
2      Abuja               Nigeria
3      Accra                 Ghana
4      Adana                Turkey

print(global_data_df.head())

   year  avg_temp
0  1750      8.72
1  1751      7.98
2  1752      5.78
3  1753      8.39
4  1754      8.47

global_data_df.describe()

	year	avg_temp
count	266.000000	266.000000
mean	1882.500000	8.369474
std	76.931788	0.584747
min	1750.000000	5.780000
25%	1816.250000	8.082500
50%	1882.500000	8.375000
75%	1948.750000	8.707500
max	2015.000000	9.830000

checking whether duplicate values exist

listOFDuplicate=city_list_df[city_list_df.duplicated()]
print(listOFDuplicate.count())

city       0
country    0
dtype: int64

listOFDuplicate=global_data_df[global_data_df.duplicated()]
print(listOFDuplicate.count())

year        0
avg_temp    0
dtype: int64

listOFDuplicate=city_data_df[city_data_df.duplicated()]
print(listOFDuplicate.count())

year        0
city        0
country     0
avg_temp    0
dtype: int64

checking wheter NaN or empty cells exist

listOfNanRows = city_data_df[city_data_df.isnull().any(axis=1)]
print(listOfNanRows.count())
#print(city_data_df[city_data_df.isnull()].count())

year        2547
city        2547
country     2547
avg_temp       0
dtype: int64

listOfNanRows = global_data_df[global_data_df.isnull().any(axis=1)]
print(listOfNanRows)

Empty DataFrame
Columns: [year, avg_temp]
Index: []

listOfNanRows = city_list_df[city_list_df.isnull().any(axis=1)]
print(listOfNanRows)

Empty DataFrame
Columns: [city, country]
Index: []

#print(city_data_df[city_data_df['avg_temp'].isnull()].head())
print(city_data_df.empty)

False

print(global_data_df.empty)

False

print(city_list_df.empty)

False

get the data of the citites in the country Turkey

citiesTurkey=city_data_df[city_data_df['country']=='Turkey']
print(citiesTurkey)

       year   city country  avg_temp
659    1791  Adana  Turkey     21.62
660    1792  Adana  Turkey     18.57
661    1793  Adana  Turkey     18.44
662    1794  Adana  Turkey     18.82
663    1795  Adana  Turkey     18.68
...     ...    ...     ...       ...
28160  2009  Izmir  Turkey     18.41
28161  2010  Izmir  Turkey     19.27
28162  2011  Izmir  Turkey     17.60
28163  2012  Izmir  Turkey     18.54
28164  2013  Izmir  Turkey     19.21

[1285 rows x 4 columns]

get the data of the city Ankara

myCityAnkara=citiesTurkey[citiesTurkey["city"]=="Ankara"]
print(myCityAnkara)

      year    city country  avg_temp
3321  1755  Ankara  Turkey      9.62
3322  1756  Ankara  Turkey     10.99
3323  1757  Ankara  Turkey     10.82
3324  1758  Ankara  Turkey      8.33
3325  1759  Ankara  Turkey      9.93
...    ...     ...     ...       ...
3575  2009  Ankara  Turkey     11.63
3576  2010  Ankara  Turkey     13.27
3577  2011  Ankara  Turkey     10.39
3578  2012  Ankara  Turkey     11.45
3579  2013  Ankara  Turkey     13.27

[259 rows x 4 columns]

myCityAnkara.describe()

	year	avg_temp
count	259.000000	259.000000
mean	1884.000000	10.394431
std	74.911058	0.703879
min	1755.000000	8.330000
25%	1819.500000	9.930000
50%	1884.000000	10.390000
75%	1948.500000	10.820000
max	2013.000000	13.270000

get the NaN values and replace it with the mean value

ankaraistOfNanRows = myCityAnkara[myCityAnkara.isnull().any(axis=1)]

myCityAnkara['avg_temp'] = myCityAnkara['avg_temp'].fillna((myCityAnkara['avg_temp'].mean()))
resultNA = myCityAnkara[myCityAnkara.isnull().any(axis=1)]
print(myCityAnkara.count())

year        259
city        259
country     259
avg_temp    259
dtype: int64

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  """Entry point for launching an IPython kernel.

print(myCityAnkara.head())
print(myCityAnkara["avg_temp"].count())

      year    city country  avg_temp
3321  1755  Ankara  Turkey      9.62
3322  1756  Ankara  Turkey     10.99
3323  1757  Ankara  Turkey     10.82
3324  1758  Ankara  Turkey      8.33
3325  1759  Ankara  Turkey      9.93
259

import matplotlib.pyplot as plt
import numpy as np
import matplotlib.pylab as plb
from matplotlib.ticker import FormatStrFormatter

plt.close("all")

descriptive statistics of this city Ankara temperature changes

maxAnkara=myCityAnkara["avg_temp"].max()
minAnkara=myCityAnkara["avg_temp"].min()
print(maxAnkara)
print(minAnkara)

13.27
8.33

meanmyCityAnkara=myCityAnkara["avg_temp"].mean()
print(meanmyCityAnkara)

10.394430894308941

standardmyCityAnkara=myCityAnkara["avg_temp"].std()
print(standardmyCityAnkara)

0.7038792574102576

visualisation of the city temperature without moving average

#myCityAnkaraBarplot = pd.Series(myCityAnkara["avg_temp"], index=range(1755,2014))
lengthOfMyCityAnkara=myCityAnkara["avg_temp"].count()
dAnkara = {'year': range(1755,2014), 'avg_temp': myCityAnkara["avg_temp"].values,
           "mean":[meanmyCityAnkara for i in range(lengthOfMyCityAnkara)],
           "standard deviation":[standardmyCityAnkara for i in range(lengthOfMyCityAnkara)]}
myCityAnkaraBarplot=pd.DataFrame(data=dAnkara)
fig = plt.figure(figsize=(16,10), dpi=600)
#plt.gca().yaxis.set_major_formatter(FormatStrFormatter('%d '))

plt.plot(myCityAnkaraBarplot["year"], myCityAnkaraBarplot["avg_temp"], color="navy",linewidth=2)
plt.plot(myCityAnkaraBarplot["year"], myCityAnkaraBarplot["mean"], color="darkorange",linewidth=3)
        
#plt.bar(myCityAnkaraBarplot["year"], myCityAnkaraBarplot["avg_temp"], color ='maroon',width = 0.4)
plt.title("Ankara temperature changes", fontsize=18)
plt.legend(["Ankara","Mean"], loc ="upper left",fontsize=14)
plt.xlabel('Years', fontsize=16)
plt.ylabel('Temperature $C^\circ$', fontsize=16)
plt.show()

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/cbook/__init__.py:1402: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  x[:, None]
/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/axes/_base.py:276: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  x = x[:, np.newaxis]
/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/axes/_base.py:278: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  y = y[:, np.newaxis]

calculation of moving averages

import numpy as np
print(myCityAnkara.shape[0])
resultMovingAvgwithTen = np.array([])
for i in range(myCityAnkara.shape[0]-10):
    arr = np.array([])    
    for j in range(i,10+i):
        avgTemp=myCityAnkara.iloc[j]['avg_temp']
        arr=np.append(arr,avgTemp)      
    meanValuesWithTenYears=np.mean(arr)
    resultMovingAvgwithTen=np.append(resultMovingAvgwithTen,meanValuesWithTenYears)

print(len(resultMovingAvgwithTen))
yearsArr=np.array([])

for i in range(0,myCityAnkara.shape[0]-10,10):
    yearsArr=np.append(yearsArr,myCityAnkara.iloc[i])
    

259
249

columns = {'avg_temp':resultMovingAvgwithTen }
resultsWithTenYears = pd.DataFrame(data=columns)
print(resultsWithTenYears.head())

    avg_temp
0  10.054772
1  10.132215
2  10.072659
3  10.030102
4  10.236545

visualisation of the city temperature with moving average

ankaraTenYearsMovingAvgPlot = pd.Series(resultMovingAvgwithTen, index=range(1765,2014))

#ax = ankaraTenYearsMovingAvgPlot.plot(colormap='jet',  title='temperature changes in Ankara with moving average')
#ax.set_xlabel("Years")
#ax.set_ylabel("Moving Averages with 10 Years")

fig = plt.figure(figsize=(16,10), dpi=600)
#plt.gca().yaxis.set_major_formatter(FormatStrFormatter('%d $C^\circ$'))

MovingAvgAnkara = {'year': range(1765,2014), 
                   'avg_temp': resultsWithTenYears["avg_temp"].values,                    
                  }
myCityAnkaraBarplot=pd.DataFrame(data=MovingAvgAnkara)
plt.plot(myCityAnkaraBarplot["year"], MovingAvgAnkara["avg_temp"], color="navy",linewidth=2)
#plt.bar(myCityAnkaraBarplot["year"], myCityAnkaraBarplot["avg_temp"], color ='maroon',width = 0.4)
plt.title("Ankara temperature changes", fontsize=18)
plt.legend(["Ankara","Mean"], loc ="upper left",fontsize=14)
plt.xlabel('Years', fontsize=16)
plt.ylabel('Moving Avg Temp $C^\circ$', fontsize=16)
plt.show()

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/cbook/__init__.py:1402: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  x[:, None]
/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/axes/_base.py:276: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  x = x[:, np.newaxis]

print(global_data_df.head())

   year  avg_temp
0  1750      8.72
1  1751      7.98
2  1752      5.78
3  1753      8.39
4  1754      8.47

descriptive statistics of global data

maxGlobal=global_data_df["avg_temp"].max()
minGlobal=global_data_df["avg_temp"].min()
meanOfGlobal=global_data_df["avg_temp"].mean()
standarDeviationGlobal=global_data_df["avg_temp"].std()
print(f"max {maxGlobal}")
print(f"min {minGlobal}")
print(f"meanOfGlobal: {meanOfGlobal}")
print(f"standard deviation: {standarDeviationGlobal}")

max 9.83
min 5.78
meanOfGlobal: 8.369473684210526
standard deviation: 0.5847474097994193

visualisation of global temperature without moving average

ig = plt.figure(figsize=(16,10), dpi=600)
#plt.gca().yaxis.set_major_formatter(FormatStrFormatter('%d $C^\circ$'))
lengthOfGlobal=global_data_df["avg_temp"].count()

print(lengthOfGlobal)
MovingAvgGlobal = {'year': range(1750,2016), 
                   'avg_temp': global_data_df["avg_temp"],
                   "mean":[meanOfGlobal for i in range(lengthOfGlobal)]
                  }
globalTenYearsMovingAveragePlot=pd.DataFrame(data=MovingAvgGlobal)
plt.plot(globalTenYearsMovingAveragePlot["year"], globalTenYearsMovingAveragePlot["avg_temp"], color="darkred",linewidth=2)
plt.plot(globalTenYearsMovingAveragePlot["year"], globalTenYearsMovingAveragePlot["mean"], color="darkorange",linewidth=3)
#plt.bar(myCityAnkaraBarplot["year"], myCityAnkaraBarplot["avg_temp"], color ='maroon',width = 0.4)
plt.title("Global temperature changes", fontsize=18)
plt.legend(["Global","Mean"], loc ="upper left",fontsize=14)
plt.xlabel('Years', fontsize=16)
plt.ylabel('Moving average temperature $C^\circ$', fontsize=16)
plt.show()

266

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/cbook/__init__.py:1402: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  x[:, None]
/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/axes/_base.py:276: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  x = x[:, np.newaxis]
/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/axes/_base.py:278: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  y = y[:, np.newaxis]

visualisation of global temperature with moving average

print(global_data_df.shape[0])
resultMovingAvgwithTenGlobal = np.array([])
for i in range(global_data_df.shape[0]-10):
    arr = np.array([])    
    for j in range(i,10+i):
        avgTemp=global_data_df.iloc[j]['avg_temp']
        arr=np.append(arr,avgTemp)      
    meanValuesWithTenYearsGlobal=np.mean(arr)
    resultMovingAvgwithTenGlobal=np.append(resultMovingAvgwithTenGlobal,meanValuesWithTenYearsGlobal)

print(len(resultMovingAvgwithTenGlobal))
yearsArr=np.array([])

for i in range(0,global_data_df.shape[0]-10,10):
    yearsArr=np.append(yearsArr,global_data_df.iloc[i])

266
256

#globalTenYearsMovingAvgPlot = pd.Series(resultMovingAvgwithTenGlobal, index=range(1760,2016))
"""
gl = globalTenYearsMovingAvgPlot.plot(colormap='jet',  title='temperature changes in global with moving average (1750-2015)')
gl.set_xlabel("Years")
gl.set_ylabel("Moving Averages with 10 Years")
"""
fig = plt.figure(figsize=(16,10), dpi=600)
#plt.gca().yaxis.set_major_formatter(FormatStrFormatter('%d $C^\circ$'))

MovingAvgGlobal = {'year': range(1760,2016), 
                   'avg_temp': resultMovingAvgwithTenGlobal,                  
                  }
globalTenYearsMovingAveragePlot=pd.DataFrame(data=MovingAvgGlobal)
plt.plot(globalTenYearsMovingAveragePlot["year"], globalTenYearsMovingAveragePlot["avg_temp"], color="darkred",linewidth=2)
#plt.bar(myCityAnkaraBarplot["year"], myCityAnkaraBarplot["avg_temp"], color ='maroon',width = 0.4)
plt.title("Global temperature changes", fontsize=18)
plt.legend(["Global"], loc ="upper left",fontsize=14)
plt.xlabel('Years', fontsize=16)
plt.ylabel('Moving average temperature $C^\circ$', fontsize=16)
plt.show()

newDataSimilartoAnakarYear=global_data_df[(global_data_df["year"] >=1755) & (global_data_df["year"] <=2013)]
print(newDataSimilartoAnakarYear.head())

   year  avg_temp
5  1755      8.36
6  1756      8.85
7  1757      9.02
8  1758      6.74
9  1759      7.99

print(newDataSimilartoAnakarYear.shape[0])
resultMovingAvgwithTenGlobal1 = np.array([])
for i in range(newDataSimilartoAnakarYear.shape[0]-10):
    arr = np.array([])    
    for j in range(i,10+i):
        avgTemp=newDataSimilartoAnakarYear.iloc[j]['avg_temp']
        arr=np.append(arr,avgTemp)      
    meanValuesWithTenYearsGlobal=np.mean(arr)
    resultMovingAvgwithTenGlobal1=np.append(resultMovingAvgwithTenGlobal1,meanValuesWithTenYearsGlobal)

print(len(resultMovingAvgwithTenGlobal1))
yearsArr=np.array([])

for i in range(0,newDataSimilartoAnakarYear.shape[0]-10,10):
    yearsArr=np.append(yearsArr,global_data_df.iloc[i])

259
249

"""
globalTenYearsMovingAvgPlot1 = pd.Series(resultMovingAvgwithTenGlobal1, index=range(1765,2014))
gl1 = globalTenYearsMovingAvgPlot1.plot(colormap='jet',  title='temperature changes in global with moving average (1755-2013)')
gl1.set_xlabel("Years")
gl1.set_ylabel("Moving Averages with 10 Years")
"""
fig = plt.figure(figsize=(16,10), dpi=600)

MovingAvgGlobal = {'year': range(1765,2014), 
                   'avg_temp': resultMovingAvgwithTenGlobal1,                  
                  }

MovingAvgAnkara = {'year': range(1765,2014), 
                   'avg_temp': resultsWithTenYears["avg_temp"].values,                    
                  }

myCityAnkaraBarplot=pd.DataFrame(data=MovingAvgAnkara)
plt.plot(myCityAnkaraBarplot["year"], MovingAvgAnkara["avg_temp"], color="navy",linewidth=2)


globalTenYearsMovingAveragePlot=pd.DataFrame(data=MovingAvgGlobal)
plt.plot(globalTenYearsMovingAveragePlot["year"], globalTenYearsMovingAveragePlot["avg_temp"], color="darkred",linewidth=2)

plt.title("Ankara vs Global temperature changes", fontsize=18)
plt.legend(["Ankara","Global"], loc ="upper left",fontsize=14)
plt.xlabel('Years', fontsize=16)
plt.ylabel('Moving average temperature $C^\circ$', fontsize=16)
plt.show()

/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/cbook/__init__.py:1402: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  x[:, None]
/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/axes/_base.py:276: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  x = x[:, np.newaxis]
/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/matplotlib/axes/_base.py:278: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  y = y[:, np.newaxis]

globalvsAnkara = pd.DataFrame(columns = ['Ankara AVG_temp', 'Global AVG_temp'])
globalvsAnkara['Ankara AVG_temp'] = myCityAnkaraBarplot["avg_temp"]

globalvsAnkara['Global AVG_temp'] = globalTenYearsMovingAveragePlot["avg_temp"]

globalvsAnkara.head()

	Ankara AVG_temp	Global AVG_temp
0	10.054772	8.143
1	10.132215	8.132
2	10.072659	8.088
3	10.030102	8.008
4	10.236545	8.012

ank = globalvsAnkara['Ankara AVG_temp'].tolist()
glb = globalvsAnkara['Global AVG_temp'].tolist()
year = globalvsAnkara.index.values.tolist()
#ymax = max(male)
#ymin = min(female)
fig = plt.figure(figsize=(16,10), dpi=600)
plt.plot(year, ank, marker=".", color="navy",linewidth=3)
plt.plot(year, glb, marker=".", color="darkred",linewidth=3)
plt.fill_between(year, ank, glb, color="orange", alpha="0.4")
plt.title("Ankara vs Global temperature changes", fontsize=18)
plt.legend(["Ankara", "Global"], loc ="upper left",fontsize=14)
plt.xlabel('Years', fontsize=16)
plt.ylabel('Moving average temperature $C^\circ$', fontsize=16)
plt.show()